Exposing materials to ionizing radiation is a reliable means of sterilization and accelerated aging but can lead to atomistic-level chemical reactions causing undesirable macroscale degradation. Quantum-based molecular dynamics (QMD) can predict these radiation-induced reaction cascades, but obtaining reliable statistics requires many independent trajectories. We develop a graph-based analysis to automatically identify chemical changes to molecular materials sampled in QMD simulations and apply this to predict characteristics of radiation damage in model silicone polymers. Large ensembles of QMD simulations are used to predict the initial reaction cascades that follow from primary knock-on atom radiation events. An algorithm to identify all topologically unique backbone structures sampled across hundreds of simulations is developed that provides a basis for quantifying network alterations such as formation of cross-links and chain scissions. The analysis framework also provides a means to assess concurrent small molecule production, yielding a rich data set linking network changes with evolution of off-gassing species. Covariance analysis reveals empirical relationships between small molecule production and specific topological features that are difficult to assess experimentally. Distinguishing characteristics of radiation coupling to different parts of the silicone are explored.

Graph-based identification of chemical damage in reactive simulations of irradiated molecular materials

ACS Spring Meeting 2021

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.